PDF《Asia》

i.e., Kosa is frequently observed in DFY. The strong wind frequency is also high in DFY and low in DNY. [11] A clear positive correlation was found between dust outbreak frequencies and strong wind frequencies in year- to-year variations. However, the month of April was an exception in both

1995 and 1998. The strong wind frequency was as high as in DFY in 1995;

nevertheless, the dust Figure 1. Total number of days for which Kosa (i.e., yellow sand) events were observed at

123 observatories for each year in Japan. For example, the total number of Kosa days is five for any day when Kosa is observed at five observatories in one day. The result of

2002 includes data until May

12 of that year. (From a report for the press by Japan Meteorological Agency;

http://www.kishou.go.jp/ press/0204/15a/kosa.pdf). Figure 2. Maximum monthly dust outbreak frequency during Jan.

1993 to Jun. 2002. The box indicates the analysis region. Thick solid lines indicate country boundaries. Thin solid lines indicate 1,500 m and 3,000 m topography contours. Gray shading indicates the sea and lakes. Figure 3. Monthly dust outbreak frequency and strong wind frequency from Jan.

1993 to Jun.

2002 (white bars and white circles) and the same from Jan.

2000 to Jun.

2002 (black bars and black dots). Figure 4. Monthly dust outbreak frequency and strong wind frequency from Jan.

1993 to Jun. 2002. The black bar chart and line graph with black dots indicate dust outbreak frequency and strong wind frequency. ASC

2 -

2 KUROSAKI AND MIKAMI: RECENT FREQUENT DUST EVENTS outbreak frequency was almost equal to the average of the DNY. In contrast, the dust outbreak frequency was some- what high in 1998, although the strong wind frequency was almost equal to the average of the DNY. The dust phenom- ena in exceptional years should be discussed in a future work, because this study focuses on discussions of dust outbreaks and surface winds. 4.3. Months That Dust Outbreaks Increased [12] The black bar chart and line graph with black dots in Figure

3 indicate the frequencies of monthly dust outbreaks and strong winds for DFY (2000C2002). Although the most frequent months for dust outbreaks are March, April, and May in a 10-year average (white bar chart), remarkable increases of dust outbreak frequencies occurred in March and April, while no increase was observed in May. The strong wind frequencies also exhibit the same tendency as the dust outbreak, i.e., increases only in March and April. 4.4. Spatial Distribution of Dust Outbreaks [13] Figures 5a and 5b illustrate the spatial distribution of the dust outbreak frequencies for March and April in DNY (1993C1999) and DFY (2000C2002). Areas of frequent dust outbreaks can be seen in the western part of the analysis region in DNY (Figure 5a), i.e., around southern Mongolia (about

44 N,

105 E), the Badain Jaran Desert (about

42 N,

102 E), and the western Loess Plateau (about

40 N,

105 E). The dust outbreak frequencies were low at most observatories in the eastern part of the analysis region, i.e., around the North China Plain (about

40 N,

115 E), northeastern China (about

45 N,

125 E), and the Korean Peninsula (about

38 N,

127 E). In contrast, the region of frequent dust outbreaks expanded to the east in DFY (Figure 5b), and dust outbreaks frequently occurred at many observatories in the eastern region. Although the area of frequent dust outbreaks expanded little into the western region,dustoutbreakfrequencyincreasedateachobservatory. 4.5. Spatial Difference in Dust Outbreaks with Strong Winds [14] Section 4.2 described the correlation between dust outbreaks and strong winds in the year-to-year variations;